At CodingCops, I lead QA strategy across four concurrent AI and consumer products and lead a 12-member QA team. I bring shift-left and risk-based testing into Agile delivery.
I designed validation frameworks for semantic search and conversational AI, evaluating 1,200+ prompts and cataloging 80+ hallucination patterns. This work improved chatbot response accuracy from 78% to 92%.
I co-defined model release benchmarks with ML engineers, increasing semantic-search relevance to 94% across 10,000+ test queries. I also validated RBAC and ABAC controls, including RAG, guardrails, and PII filtering.
At CodingCops, I architected Playwright and Appium automation suites that reduced regression execution time from 12 hours to 3 hours. Previously, at Speridian Technologies, I developed and maintained a regression suite covering 500+ test cases and performed API, ETL, and performance testing.
